Cystathionine B-Synthase Polymorphisms and Hyperhomocysteinaemia: an Association Study

ARTICLE Cystathionine b-synthase polymorphisms and hyperhomocysteinaemia: an association study

Karin JA Lievers1, Leo AJ Kluijtmans1, Sandra G Heil1, Godfried HJ Boers2, Petra Verhoef4, Martin den Heijer3, Frans JM Trijbels1 and Henk J Blom*,1

1Laboratory of Pediatrics and Neurology, University Medical Center Nijmegen, The Netherlands; 2Department of Internal Medicine, University Medical Center Nijmegen, The Netherlands; 3Department of Internal Medicine, Division of Endocrinology, University Medical Center Nijmegen, The Netherlands; 4Wageningen Center for Food Sciences and Division of Human Nutrition and Epidemiology, Wageningen University, Wageningen, The Netherlands

Hyperhomocysteinaemia is generally accepted as an independent and graded risk factor for both arterial occlusive disease and venous thrombosis. The only way of homocysteine degradation is conversion to cysteine in the transsulfuration pathway in which the regulating step is catalysed by cystathionine b- synthase (CBS). Mild impairment of CBS function could therefore affect homocysteine concentration, in particular after methionine loading, and consequently cardiovascular disease (CVD) risk. We analysed two silent polymorphisms and one short tandem repeat in the CBS gene (ie 699C?T, 1080C?T and – 5697 (GT) STR) as genetic markers potentially in linkage disequilibrium with a functional polymorphism. We assessed their association with fasting and post-methionine load homocysteine in 190 patients with arterial occlusive disease, and in 381 controls. No differences in CBS genotype frequencies between cases and controls were found, nor was a particular CBS genotype associated with an elevated risk of arterial occlusive disease. Although we did find a high rate of linkage disequilibrium between the two single nucleotide polymorphisms and the GT STR, none of the genotypes defined by the three CBS variants studied showed an association with elevated fasting, post-load or increase upon methionine loading homocysteine concentrations. In conclusion, we did not find any indication that genetic variation in the CBS gene is associated with increased homocysteine concentrations. European Journal of Human Genetics (2003) 11, 23 – 29. doi:10.1038/sj.ejhg.5200899

Keywords: cystathionine b-synthase (CBS); homocysteine; polymorphism; cardiovascular disease; association study; linkage disequilibrium

Introduction methionine metabolism, may either be irreversibly Established risk factors for cardiovascular disease (CVD) degraded to cysteine via cystathionine in the transsulfura- include elevated blood cholesterol, hypertension, diabetes tion pathway or remethylated to methionine. The mellitus and tobacco smoking. A modest elevation of plas- remethylation by methionine synthase requires 5-methyl- ma homocysteine concentration, commonly referred to as tetrahydrofolate for methyl donation and vitamin B12 for hyperhomocysteinemia, is generally,1,2 though not univer- methyl transport, while the remethylation by betaine- sally3,4 accepted as an independent and graded risk factor homocysteine-methyltransferase requires betaine as methyl for both arterial occlusive diseases and venous thrombo- donor. The regulating step in the transsulfuration pathway sis.5,6 Homocysteine, an intermediate sulfur amino acid in is catalysed by cystathionine b-synthase (CBS), a pyridoxal 5’-phosphate-dependent enzyme. The inherited metabolic *Correspondence: Dr HJ Blom, Laboratory of Pediatrics and Neurology, disorder CBS deﬁciency is acknowledged to be the most University Medical Center Nijmegen, PO Box 9101, 6500 HB, Nijmegen, frequent cause of homocystinuria in humans.7 The Netherlands. Tel: +31 24 3613469; Fax: +31 24 3618900; 8 E-mail: [email protected] In 1985, Boers et al postulated heterozygosity for CBS Received 4 June 2002; revised 7 August 2002; accepted 26 August 2002 deﬁciency as a major cause of hyperhomocysteinaemia in CBS association study KJA Lievers et al 24

CVD patients. This was based on the decreased CBS activ- be in linkage disequilibrium with a functional polymorph- ities they found in extracts of cultured fibroblasts of ism. We analysed two silent single nucleotide hyperhomocysteinaemic vascular disease patients and their polymorphisms (699C?T, 1080C?T) and a short tandem increased post-methionine loading homocysteine concen- repeat (STR) – 5697(GT),22 in the CBS gene in 190 vascular trations, both resembling those of obligate heterozygotes disease patients and in 381 population-based controls. We for CBS deficiency. Similar results were described by Clarke report the allele frequencies and genotype distributions of et al.9 Both studies however, do not reconcile with an these polymorphic variants in both study groups, and earlier observation by Mudd et al,10 who described a similar describe their association with fasting, post-methionine incidence of heart attacks or strokes in parents and grand- load and delta (ie increase upon methionine loading) plas- parents of homocystinuric children compared with ma homocysteine concentrations. Also, the extent of relatives of children with either new-mutation achondro- linkage disequilibrium between the several CBS variants plastic dwarfism or with impaired phenylalanine are calculated. metabolism. Furthermore, the finding of decreased CBS activities in cultured fibroblasts from hyperhomocysteinae- Materials and methods mic vascular disease patients in the range of obligate Study population heterozygotes could not be reproduced by us,11,12 and still We studied 190 patients with coronary, peripheral or cere- needs clarification. In addition to fasting homocysteine bral vascular disease; a total of 130 cases were recruited concentrations, also post-methionine load homocysteine from a cohort of patients who underwent coronary angio- concentrations were found to be correlated among family graphy in the Zuiderziekenhuis Hospital in Rotterdam, members of patients with hyperhomocysteinaemia and The Netherlands.18 Reasons for angiography were myocar- vascular disease.13 – 15 So, it can be concluded that hyper- dial infarction or angina pectoris and cases were defined homocysteinaemia is, at least partially, genetically based. as those having 590% occlusion in one and 540% occlu- Still, elevated post-methionine load homocysteine concen- sion in one additional coronary artery. The other 60 trations are associated with increased CVD risk,16 – 18 but patients had documented premature cardiovascular disease its genetic determinants remain obscure. of which 10 patients had suffered from myocardial infarc- Over 100 mutations have been described in the CBS gene tion, 32 from cerebral arterial occlusive disease and 18 in classical homocystinuria patients.19 A 68 bp insertion in from peripheral arterial occlusive disease.12 exon 8, first reported as a causal mutation in an Italian The control group consisted of 381 controls, of whom patient, was found to be polymorphic and not to be asso- 101 were included from the general population in Rotter- ciated with plasma homocysteine concentrations.20,21 No dam and comprised subjects with no history of other potentially functional polymorphisms associated with cardiovascular disease18 and 280 were recruited from a elevated homocysteine levels were detected in the CBS gene general practice in The Hague, The Netherlands to take part in the analyses of the genetic basis of homocystinuria. in a health survey.25 However, most studies performed thus far only examined the coding region of the CBS gene, leaving the non-coding Biochemical parameters regions unexplored. After an overnight fast, blood was collected in EDTA tubes Very recently, we described an inverse correlation for measurement of fasting total homocysteine. The between the number of tandem repeats defined by a EDTA-blood samples were placed on ice immediately, and 31 bp VNTR (variable number of tandem repeats),22 and centrifuged for 10 min at 3000 g with minimal delay. All CBS enzyme activities in fibroblasts.23 Furthermore, a posi- individuals were subjected to a standardised methionine tive association was observed between the number of repeat loading test.26 Briefly, L-methionine (0.1 g/kg body weight) units and post-methionine load plasma homocysteine was dissolved in 200 mL orange juice and administered concentrations.23 orally. All study subjects received a standardised low protein Genetic association studies are widely used for the iden- breakfast and luncheon, and were asked not to consume tification of candidate genes of complex, ie non-mendelian any milk-containing beverages. After 6 h, another blood traits.24 In such studies, the association is investigated sample was drawn for assessment of the post-methionine between a phenotype and genomic markers close to or, load homocysteine concentration. Total homocysteine was preferably, within the gene of interest. Such a marker in measured in plasma using a high-performance liquid chro- itself is not necessarily functionally related to the pheno- matography (HPLC) procedure, with reverse phase type but may be in linkage disequilibrium with a separation and fluorescence detection, as described by Te functional variant. Poele-Pothoff et al.27 All homocysteine measurements were In the present study, we investigated the potential invol- conducted at the University Medical Center Nijmegen, The vement of genetic variants in non-coding regions or of Netherlands, and fasting and post-load plasma homocys- silent mutations in the CBS gene in hyperhomocysteinae- teine levels were obtained from 555 and 525 individuals, mia and vascular disease, as possible markers which may respectively.

European Journal of Human Genetics CBS association study KJA Lievers et al 25

Detection of CBS polymorphisms used in all analyses. Differences in genotype distributions, Genomic DNA was isolated from buffy coats of EDTA blood allele frequencies and gender were assessed by Pearson’s w2 by a standard method,28 and stored at 48C until analysis. analysis. Differences in continuous variables between cases Genomic DNA was available from 561 individuals. and controls were assessed by Student’s t-test, adjusted for age and gender by means of linear regression analysis. 699C?T transition One-way analysis of variance (ANOVA) was used to assess This polymorphism does not create a restriction enzyme the differences of continuous variables between different recognition site, and was therefore analysed by Primer genotypes, followed by Bonferroni-corrected t-tests. Odds Introduced Restriction Analysis. A mutagenic sense oligonu- ratios (OR) and 95% confidence intervals (95% CI) as esti- cleotide (5’-CAGCAACCCCCTGGCTCAGT-3’) introduces a mates of the relative risk for arterial occlusive disease for RsaI site in the 699C allele and together with an antisense the different CBS genotypes, were calculated with logistic oligonucleotide (5’-TTATCGTTTGTGTCCCGTACCG-3’)a regression analysis. We also investigated the association genomic DNA fragment of 287 bp was amplified using between these three genetic variants and the 31 bp VNTR, approximately 100 ng genomic DNA and 100 ng of both of which alleles 19 and 21 were shown to be associated oligonucleotides in a standard PCR buffer (10 mM Tris- with increased post-methionine load homocysteine concen- 23 HCl, pH 8.3; 50 mM KCl) containing 2 mM MgCl2 and trations. The haplotype frequencies and linkage 29 200 mM dNTPs. PCR parameters were as follows: 3 min initi- disequilibria were estimated by the EH program. D’ al denaturation at 928C, followed by 35 cycles of 928C/60 s, reflects the extent of the linkage disequilibrium, which is 648C/60 s, 728C/30 s, and a final extension of 7 min at reported as the ratio between the actual value of D (ie the 728C. After digestion with RsaI and resolution of the frag- departure from linkage disequilibrium) and the maximum ments on a 20% polyacrylamide gel, the 699 CC genotype value it can have for the given allele frequencies. The sign results in fragments of 171, 92, 20 and 4 bp, whereas the in front of the coefficients indicates whether the rare alleles 699 TT genotype shows fragments of 171, 112 and 4 bp. are associated (+) or the rare allele is associated with the frequent allele (7). All P-values reported are two-tailed, 1080C?T transition and statistical significance was accepted at P50.05. A genomic DNA fragment of 88 bp was amplified by PCR using both 100 ng forward (5’-CTGGCAGCACGGTGGC- Results GG-3’) and 100 ng reverse oligonucleotides (5’-CGCACTG- Table 1 depicts age, gender, and total plasma homocysteine AGTCGGGCAGAATG-3’), and approximately 100 ng geno- concentrations (ie fasting, post methionine load, and net mic DNA in a standard PCR buffer (10 mM Tris-HCl, pH 8.3; increase upon methionine loading) of cases and controls.

50 mM KCl) containing 2 mM MgCl2 and 200 mM dNTPs. All Age, expressed as the geometric mean, and the percentage samples were cycled 35 times: 1 min /928C denaturation, males did not differ between the two groups. A higher fast- 1 min/558C annealing, 30 s/728C extension, preceded by an ing homocysteine (+7.5%; P=0.011) and post-methionine initial denaturation of 3 min/928C and followed by a final loading homocysteine concentration (+5.6%; P=0.045) was extension of 7 min/728C. The PCR fragment was analysed seen in cases compared with controls. Cases and controls by BstUI restriction enzyme analysis followed by separation showed a comparable net increase upon methionine load- on a 2% agarose gel. The 1080 TT genotype results in an ing (ie post-load minus fasting homocysteine uncut fragment of 88 bp, the 1080 CC genotype in two frag- concentration). ments of 55 and 33 bp, whereas the heterozygous CT Both 699C?T, 1080C?T as well as the GT STR genotype genotype is displayed by three fragments (88, 55, and 33 bp). distributions were in Hardy-Weinberg equilibrium (data not shown). Genotype distributions for each of the three CBS 75697 (GT) STR polymorphisms were similar in cases and controls (Table A PCR was performed with a 5’FAM labelled forward primer 2, left side). The allele frequency of the 699T allele was (5’-GAGCTGAAATCACACCACT-3’) and a reverse primer (5’- 0.37 and 0.35 in patients and controls, respectively GTTTTTACTACATTTGCTCCC-3’) in a standard PCR buffer (w2=0.30, P=0.58); the 1080T allele had an allele frequency 2 containing 2 mM MgCl2 and using an annealing temperature of 0.32 in patients and of 0.37 in controls (w =2.89, of 588C. The 193 – 211 bp fluorescently labelled PCR products P=0.09). In our study population, we identified 10 different were separated on a 4% denaturating polyacrylamide gel alleles defined by the highly polymorphic GT STR (which using the ABI Prism 377, according to the manufacturer’s were named after the number of repeat units) that led to recommendations. Data-analysis was performed using the the observation of 22 different genotype combinations Genescan and Genotyper Software (Perkin Elmer). (Table 2). CBS genotypes were investigated as risk factors for arter- Statistical analysis ial occlusive disease. The differences in CBS genotype Plasma homocysteine concentrations showed positive skew- frequencies between the case and control group were not ness, therefore log-transformed homocysteine levels were statistically significant (Table 2). Odds ratios for the 699TT

European Journal of Human Genetics CBS association study KJA Lievers et al 26

Table 1 Characteristics and plasma homocysteine concentration (tHcy) in cases and controls Cases Controls (n=190) (n=381) pa

Age (years) 49.0+10.2 50.8+11.9 0.11 Gender (% male) 75.3 74.3 0.80b Fasting tHcy (mmol/L) 14.4 (13.7 – 15.1) 13.4 (13.0 – 13.9) 0.011c Postload tHcy (mmol/L) 41.3 (39.4 – 43.3) 39.1 (38.0 – 40.3) 0.045c Postload increased (mmol/L) 26.1 (24.6 – 27.7) 24.7 (23.6 – 25.9) 0.13c Age is expressed as mean+SD and tHcy as geometric means (95% CI). aStudent’s t-test. bPearson w2 test. cAdjusted for age and gender. dIncrease upon methionine loading (ie postload minus fasting tHcy).

Table 2 CBS genotype distribution in cases and controls and the relationship between plasma homocysteine concentration (tHcy) and CBS genotypes (patients and controls combined) Number of Fasting Postload CBS Patients Controls tHcy tHcy genotype N (%) N (%) mmol/L mmol/L D tHcya

699CC 76 (40.0) 156 (42.9) 13.9 40.0 25.1 699CT 87 (45.8) 158 (43.4) 13.5 39.6 25.5 699TT 27 (14.2) 50 (13.7) 14.2 40.0 25.1 w2=0.41; P=0.81 P=0.60b P=0.92b P=0.91b 1080CC 93 (49.2) 150 (40.4) 14.0 40.6 25.9 1080CT 70 (37.0) 164 (44.2) 13.6 39.1 24.9 1080TT 26 (13.8) 57 (15.4) 13.5 39.7 24.4 w2=3.98; P=0.14 P=0.58b P=0.38b P=0.44b GT STR 14-14 1 (0.5) 1 (0.3) 20.1 44.3 24.2 11-16 0 1 (0.3) 11.4 31.8 20.5 12-16 1 (0.5) 0 11.6 - - 13-16 0 1 (0.3) 9.6 24.3 14.8 14-16 9 (4.8) 33 (9.6) 14.3 40.6 26.1 15-16 10 (5.3) 11 (3.2) 15.0 41.7 25.9 16-16 75 (40.1) 146 (42.7) 13.5 39.4 25.2 14-17 2 (1.1) 3 (0.9) 15.4 41.6 25.8 16-17 47 (25.1) 79 (23.1) 13.6 40.3 26.1 17-17 3 (1.6) 4 (1.2) 18.3 46.7 28.1 14-18 1 (0.5) 4 (1.2) 13.1 37.5 23.6 15-18 1 (0.5) 1 (0.3) 13.4 47.6 33.5 16-18 14 (7.5) 25 (7.3) 13.7 38.7 24.7 17-18 2 (1.1) 8 (2.3) 15.3 38.1 22.5 18-18 2 (1.1) 1 (0.3) 11.6 34.3 22.6 14-19 1 (0.5) 2 (0.6) 13.6 45.3 30.5 16-19 11 (5.9) 16 (4.7) 15.1 39.4 20.6 17-19 5 (2.7) 3 (0.9) 13.2 44.8 30.5 18-19 0 1 (0.3) 12.8 38.5 25.8 19-19 0 1 (0.3) 7.8 20.7 12.8 16-20 1 (0.5) 0 18.9 46.1 27.0 17-20 1 (0.5) 1 (0.3) 13.0 38.2 25.0 w2=17.74; P=0.67 P=0.42b P=0.78b P=0.73b tHcy expressed as geometric means (95% CI). aNet increase upon methionine loading (ie postload minus fasting tHcy). bANOVA on log transformed data.

and 699CT genotype, with the 699CC genotype as reference was studied (Table 2, right side). We combined cases and category, were 1.07 (95% CI: 0.81 – 1.40) and 1.12 (95% CI: controls to increase statistical power, since associations 0.77 – 1.64) respectively. Odds ratios for the 1080TT and between CBS genotypes and plasma homocysteine did not 1080CT genotype, with 1080CC as reference category, were differ between cases and controls (all P-values 40.14). For 1.07 (95% CI: 0.89 – 1.52) and 0.94 (95% CI: 0.54 – 1.62) the 699C?T polymorphism no differences in fasting, respectively. post-load or delta homocysteine concentrations were In addition, the association between CBS genotypes and observed between the three genotypes (Table 2). Further- fasting and post-load plasma homocysteine concentrations more, none of the genotypes deﬁned by the 1080C?T

European Journal of Human Genetics CBS association study KJA Lievers et al 27 polymorphism showed an association with elevated fasting, in the CBS gene by analysing the 699C?T and 1080C?T post-load or delta homocysteine concentration (Table 2). variants and a GT STR, which is located in the promotor The different GT STR genotypes were also not associated region, as markers of possible genetic CBS defects contribut- with a change in plasma homocysteine concentrations ing to hyperhomocysteinaemia and to the risk of arterial (Table 2). We also investigated the association between occlusive disease. No differences in genotype distributions these three genetic variants and the 31 bp VNTR, of which or allele frequencies were found between cases and controls, alleles 19 and 21 were shown to be associated with indicating that no particular CBS genotype is over-repre- increased post-methionine load homocysteine concentra- sented in arterial occlusive disease patients compared with tions.23 The positive association between alleles 19 and 21 controls. We investigated the associations between CBS and post-load homocysteine concentrations was however genotypes and plasma homocysteine concentrations, espe- not reflected in one of the genotypes defined by the three cially after methionine loading. The two silent mutations variants studied here. We examined whether alleles 19 as well as the GT STR did not show an association with and 21 were located on particular haplotypes, but found homocysteine concentrations, neither fasting, nor post- out that this was not the case; six different haplotypes were load, nor on the increase upon methionine loading. observed containing allele 19, and three containing allele Besides fasting plasma homocysteine concentrations,1,18 21 (Table 3). also post-methionine load homocysteine concentrations Linkage disequilibria were calculated between the have been associated with an increased CVD risk.16,18,30 699C?T, 1080C?T and the GT STR and all were found to Verhoef et al18 calculated that the risk of coronary artery be highly significant (P50.001; data not shown). All pair- disease increased with 20% for each 12 mmol/L increase in wise linkage disequilibria between the different CBS alleles post-load homocysteine levels. Graham et al31 reported a are depicted in Table 4. relative risk for vascular disease of 1.5 for a subject with a post-load homocysteine level above the 80th percentile of Discussion controls. Since methionine loading particularly promotes In the present study, we investigated the potential involve- the transsulfuration of homocysteine into cysteine, in ment of variants in non-coding regions or silent mutations which CBS is the regulating enzyme, it has been postulated that hyperhomocysteinaemia after methionine loading reflects the involvement of suboptimal CBS function.32 Several studies excluded the involvement of heterozygosity Table 3 Haplotypes of the CBS 31 bp VNTR alleles 19 and 21 and their relative frequencies for CBS deficiency in hyperhomocysteinaemia and premature vascular disease,12,33,34 but the results of these studies Polymorphisms Relative do not exclude an involvement of a mildly impaired CBS Haplotype 31 bp VNTR GT STR 699C?T 1080C?T frequency function. No potentially functional polymorphisms asso- A 19 15 T C 0.036 ciated with hyperhomocysteinaemia after methionine B 19 16 C C 0.074 loading were detected in the CBS gene in the analyses of C 19 16 C T 0.036 D 19 16 T C 0.448 the genetic basis of homocystinuria. However, most studies E 19 17 C T 0.036 performed thus far only examined the coding region of the F 19 17 T C 0.085 CBS gene, leaving the non-coding regions unexplored. The G 21 16 C C 0.190 effect of a potential functional polymorphism may be H 21 16 T C 0.038 I 21 17 C C 0.058 reflected in other, non-functional, polymorphisms, in case they are in linkage disequilibrium. Recently, Kruger et al35 calculated odds ratios of the 699TT and 1080CC genotypes versus the other two geno- Table 4 Linkage disequilibrium coefficients (D’) between types defined by each variant on the risk of coronary different alleles of CBS polymorphisms. The sign in front of artery disease and found decreased risks, ie protective the coefficients indicates whether the rare alleles are effects. However, individuals with the 699TT or 1080CC associated (+) or whether the rare allele is associated with genotype had similar fasting homocysteine concentrations the frequent allele (7) as the other genotypes and, unfortunately, no post-load Polymorphism Allele 699C?T 1080C?T homocysteine concentrations were available. They postulated that the 699T and 1080T alleles, may be linked to 1080C?T 70.75 7 other yet unidentified functional polymorphisms in the 75697 (GT) STR 14 0.71 70.80 15 0.25 70.35 CBS gene. These results were similar to those of De Stefano 16 0.25 70.08 et al36 who also did not find an effect of both CBS 699C?T 17 70.35 0.52 and the 1080C?T CBS variants on fasting homocysteine 18 70.47 0.67 19 70.32 70.36 concentrations. They did not examine the association between the CBS variants and post-methionine load homo-

European Journal of Human Genetics CBS association study KJA Lievers et al 28

cysteine concentrations either. Aras et al37 reported leading to elevated homocysteine concentrations and decreased post-methionine load homocysteine concentra- increased CVD risk was found. tions in individuals who were heterozygous or homozygous for the 699T allele compared with the 699CC subjects. This association became more significant Acknowledgements when individuals carrying the 844ins68 and the 1080T This study was, in part, supported by grants 93.176 and 97.071 allele were excluded. With regard to the 1080C?T poly- from the Netherlands Heart Foundation. Dr Leo AJ Kluijtmans is a post-doctoral fellow of the Netherlands Heart Foundation morphism, the 1080T allele was associated with lower (D99.023) and Dr Henk J Blom is an Established Investigator of post-methionine load homocysteine levels only when indi- the Netherlands Heart Foundation (D97.021). viduals carrying the 844ins68 and the 699T allele were excluded from the analysis. They also speculated that the 699C?T and 1080C?T CBS variants may be in linkage disequilibrium with regulatory elements that upregulate References CBS gene transcription.37 In the current study however, 1 Wald NJ, Watt HC, Law MR, Weir DG, McPartlin J, Scott JM: the two silent polymorphisms as well as the GT STR did Homocysteine and ischemic heart disease: results of a prospective study with implications regarding prevention. Arch Intern Med not show any association with homocysteine concentra- 1998; 158: 862 – 867. tions, neither fasting, nor post-load, nor on the increase 2 Ueland PM, Refsum H, Beresford SA, Vollset SE: The controversy upon methionine loading. Both etiologic and genetic over homocysteine and cardiovascular risk. Am J Clin Nutr 2000; heterogeneity may play an important role in genetic asso- 72: 324 – 332. 3 Folsom AR, Nieto FJ, McGovern PG et al: Prospective study of ciation studies. Since the populations used in the different coronary heart disease incidence in relation to fasting total studies may have a different background, it is possible that homocysteine, related genetic polymorphisms, and B vitamins: the actual range of etiologies vary across samples. Allele the Atherosclerosis Risk in Communities (ARIC) study. Circulation frequencies of the 699C?T and 1080C?T polymorphisms 1998;98: 204 – 210. 4 Brattstrom L, Wilcken DE: Homocysteine and cardiovascular were comparable between the three studies published disease: cause or effect?. Am J Clin Nutr 2000;72: 315 – 323. earlier35 – 37 and the present one, ranging from 31.8 to 5 Den Heijer M, Koster T, Blom HJ et al: Hyperhomocysteinemia as 39.4% for the 699T allele and from 32.4 to 35.5% for the a risk factor for deep-vein thrombosis. N Engl J Med 1996; 334: 759 – 762. 1080T allele. 6 Den Heijer M, Rosendaal FR, Blom HJ, Gerrits WB, Bos GM: We also investigated the association between these three Hyperhomocysteinemia and venous thrombosis: a meta-analysis. genetic variants and the 31 bp VNTR, of which the alleles Thromb Haemost 1998;80: 874 – 877. containing 19 and 21 repeat units were shown to be asso- 7 Mudd SH, Levy HL, Skovby F (ed): Disorders of transsulfuration; in Scriver CS, Beaudet AL, Sly WS, Valle D (eds): The Metabolic and ciated with increased post-methionine load homocysteine Molecular Basis of Inherited Disease. New York: McGraw-Hill, 1995; concentrations.23 However, this positive association pp 1279 – 1327. between alleles 19 and 21 and post-load homocysteine 8 Boers GHJ, Smals AG, Trijbels JMF et al: Heterozygosity for homocystinuria in premature peripheral and cerebral occlusive arterial concentrations was not reflected in one of the genotypes disease. N Engl J Med 1985; 313: 709 – 715. defined by the three variants studied here. One explanation 9 Clarke R, Daly L, Robinson K et al: Hyperhomocysteinemia: an is that alleles 19 and 21 are rare (allele frequencies of 10.5 independent risk factor for vascular disease. N Engl J Med 1991; and 2.7%, respectively) and are therefore not picked up 324: 1149 – 1155. 10 Mudd SH, Havlik R, Levy HL, McKusick VA, Feinleib M: A study of by a genetic association analysis described here. Further- cardiovascular risk in heterozygotes for homocystinuria. Am J more, we observed that alleles 19 and 21 were both not Hum Genet 1981;33: 883 – 893. located on one particular haplotype, but that different 11 Engbersen AM, Franken DG, Boers GH, Stevens EM, Trijbels FJ, haplotypes were observed containing allele 19, which was Blom HJ: Thermolabile 5,10-methylenetetrahydrofolate reductase as a cause of mild hyperhomocysteinemia. Am J Hum Genet also the case for allele 21 (Table 3). 1995;56: 142 – 150. In the present study, we detected strong linkage disequi- 12 Kluijtmans LA, van den Heuvel LP, Boers GH et al: Molecular libria between the CBS variants studied. For each pair of genetic analysis in mild hyperhomocysteinemia: a common mutation in the methylenetetrahydrofolate reductase gene is a alleles, defined by the three CBS variants, linkage disequili- genetic risk factor for cardiovascular disease. Am J Hum Genet brium coefficients (D’) were calculated (Table 4). D’ between 1996;58: 35 – 41. the 699C?T and the 1080C?T polymorphisms of 70.75 in 13 Genest JJ, McNamara JR, Upson B et al: Prevalence of familial our study is similar to that calculated by De Stephano et al36 hyperhomocyst(e)inemia in men with premature coronary artery disease. Arterioscler Thromb 1991;11: 1129 – 1136. of 70.83 in almost 800 subjects recruited from 11 European 14 Wu LL, Wu J, Hunt SC et al: Plasma homocyst(e)ine as a risk factor countries. for early familial coronary artery disease. Clin Chem 1994;40: In conclusion, we did not find any genotype, defined by 552 – 561. 15 Franken DG, Boers GH, Blom HJ, Cruysberg JR, Trijbels FJ, Hamel the three CBS variants studied here, associated with BC: Prevalence of familial mild hyperhomocysteinemia. Athero- elevated homocysteine concentrations. Although we sclerosis 1996; 125: 71 – 80. observed strong linkage between these three polymorphisms, no evidence for a common CBS polymorphism

European Journal of Human Genetics CBS association study KJA Lievers et al 29

16 Bostom AG, Jacques PF, Nadeau MR, Williams RR, Ellison RC, 28 Miller SA, Dykes DD, Polesky HF: A simple salting out procedure Selhub J: Post-methionine load hyperhomocysteinemia in for extracting DNA from human nucleated cells. Nucleic Acids Res persons with normal fasting total plasma homocysteine: initial 1988;16: 1215. results from the NHLBI Family Heart Study. Atherosclerosis 1995; 29 Xie X, Ott J: Testing linkage disequilibrium between a disease 116: 147 – 151. gene and marker loci. Am J Hum Genet 1993;53: 1107 (Abstract). 17 Kang SS, Wong PW: Genetic and nongenetic factors for moderate 30 Verhoef P, Meleady R, Daly LE, Graham IM, Robinson K, Boers hyperhomocyst(e)inemia. Atherosclerosis 1996; 119: 135 – 138. GH: Homocysteine, vitamin status and risk of vascular disease; 18 Verhoef P, Kok FJ, Kruyssen DA et al: Plasma total homocysteine, effects of gender and menopausal status. European COMAC B vitamins, and risk of coronary atherosclerosis. Arterioscler Group. Eur Heart J 1999;20: 1234 – 1244. Thromb Vasc Biol 1997;17: 989 – 995. 31 Graham IM, Daly LE, Refsum HM et al: Plasma homocysteine as a 19 Kraus JP, Janosik M, Kozich V et al: Cystathionine beta-synthase risk factor for vascular disease. The European Concerted Action mutations in homocystinuria. Hum Mutat 1999;13: 362 – 375. Project. JAMA 1997; 277: 1775 – 1781. 20 Tsai MY, Bignell M, Schwichtenberg K, Hanson NQ: High preva- 32 Finkelstein JD, Martin JJ: Methionine metabolism in mammals. lence of a mutation in the cystathionine beta-synthase gene. Distribution of homocysteine between competing pathways. J Am J Hum Genet 1996;59: 1262 – 1267. Biol Chem 1984; 259: 9508 – 9513. 21 Kluijtmans LA, Boers GH, Trijbels FJ, van Lith-Zanders HM, van 33 Kozich V, Kraus E, de Franchis R et al: Hyperhomocysteinemia in den Heuvel LP, Blom HJ: A common 844INS68 insertion variant premature arterial disease: examination of cystathionine beta- in the cystathionine beta-synthase gene. Biochem Mol Med 1997; synthase alleles at the molecular level. Hum Mol Genet 1995;4: 62: 23 – 25. 623 – 629. 22 Kraus JP, Oliveriusova J, Sokolova J et al: The human cystathio- 34 Gallagher PM, Meleady R, Shields DC et al: Homocysteine and nine beta-synthase (CBS) gene: complete sequence, alternative risk of premature coronary heart disease. Evidence for a common splicing, and polymorphisms. Genomics 1998;52: 312 – 324. gene mutation. Circulation 1996;94: 2154 – 2158. 23 Lievers KJ, Kluijtmans LA, Heil SG et al: A 31 bp VNTR in the 35 Kruger WD, Evans AA, Wang L et al: Polymorphisms in the CBS cystathionine beta-synthase (CBS) gene is associated with gene associated with decreased risk of coronary artery disease reduced CBS activity and elevated post-load homocysteine levels. and increased responsiveness to total homocysteine lowering by Eur J Hum Genet 2001;9: 583 – 589. folic acid. Mol Genet Metab 2000;70: 53 – 60. 24 Lander ES, Schork NJ: Genetic dissection of complex traits. Science 36 De Stefano V, Dekou V, Nicaud V et al: Linkage disequilibrium at 1994; 265: 2037 – 2048. the cystathionine beta synthase (CBS) locus and the association 25 Den Heijer M, Blom HJ, Gerrits WB, Rosendaal FR, Haak HL, between genetic variation at the CBS locus and plasma levels of Wijermans PW, Bos GM: Is hyperhomocysteinaemia a risk factor homocysteine. Ann Hum Genet 1998;62: 481 – 490. for recurrent venous thrombosis?. Lancet 1995; 345: 882 – 885. 37 Aras O, Hanson NQ, Yang F, Tsai MY: Inﬂuence of 699C?T and 26 Boers GHJ, Smals AG, Trijbels JMF, Leermakers AI, Kloppenborg 1080C?T polymorphisms of the cystathionine beta-synthase PW: Unique efﬁciency of methionine metabolism in premeno- gene on plasma homocysteine levels. Clin Genet 2000;58: 455 – pausal women may protect against vascular disease in the 459. reproductive years. J Clin Invest 1983;72: 1971 – 1976. 27 Te Poele-Pothoff MT, van den Berg M, Franken DG et al: Three different methods for the determination of total homocysteine in plasma. Ann Clin Biochem 1995;32: 218 – 220.

European Journal of Human Genetics